All of us regularly come across captchas when we register at a website or make a transaction. So why do websites (including our blog site) use capchas in the first place ?
Wikipedia says: A CAPTCHA is a type of challenge-response test used in computing as an attempt to ensure that the response is generated by a person.
In other words, it is a mechanism to fend off automated bots/trolls from degrading the quality of service and ensures only real people use the service.
A human rights group came up with a fancy method to the challenge-response test. Unlike the traditional method of using a computation or a scrambled text as a challenge-response test, the group uses questions whose answers are based on empathy. The Civil-Rights capchas as the group calls them have questions involving basic human rights. They intend to promote the captchas in countries which have lower importance to human rights. This serves the dual purpose of thwarting bots and at the same time also educating human rights.
This lead me into thinking of making a computer science captcha exclusively for the CS-departments. Instead of entering some random text, why not learn a new definition/concept about CS? The questions could also be related to some historic facts about computer science.
An example captcha :
3-SAT has the following complexity : a) NP-Complete b) NP-hard c) Polynomial time
Ofcourse the above options will be scrambled.
So what are the benefits? My take is that being a computer scientist, you must know the breadth of CS to some extend. The captcha question might actually lead you to some useful information you did not know. With all probabilities, the interviewer where you applied your next summer internship might actually lead you to this question.
References
http://www.wired.com/threatlevel/2012/10/empathy-captcha/
http://captcha.civilrightsdefenders.org/
http://en.wikipedia.org/wiki/CAPTCHA
I found out that research done at Stanford says that the captcha code could be easily broken and they have tried it on Visa,CNN, eBay etc. They do it by the help of removing the image background and breaking text strings into single character for easier recognition. So the point over here is if the hackers are able to read the question and have a definite list of answers for that types of questions it wont take long to break the code, but in the case where we are throwing some random strings at the user it might take long to break the code.
Source:http://phys.org/news/2011-11-stanford-outsmart-captcha-codes.html